@Override
public int run(String[] arg0) throws Exception {
// config a job and start it
Configuration conf = getConf();
Job job = new Job(conf, "Index construction..");
job.setJarByClass(IndexConstructorDriver.class);
job.setMapperClass(IndexConstructorMapper.class);
job.setReducerClass(IndexConstructorReducer.class);
job.setOutputKeyClass(Text.class);
job.setOutputValueClass(InvertedListWritable.class);
job.setMapOutputKeyClass(Text.class);
job.setMapOutputValueClass(Text.class);
job.setInputFormatClass(SequenceFileInputFormat.class);
job.setOutputFormatClass(SequenceFileOutputFormat.class);
// can add the dir by the config
FileSystem fs = FileSystem.get(conf);
String workdir = conf.get("org.joy.crawler.dir", "crawler/");
fs.delete(new Path(workdir + "indexOutput/"), true);
FileInputFormat.addInputPath(job, new Path(workdir + "content/"));
FileOutputFormat.setOutputPath(job, new Path(workdir + "indexOutput/"));
System.out.println(
"indexer starts to work, it begins to construct the index, please wait ...\n");
return job.waitForCompletion(true) ? 0 : 1;
}
public static void main(String args[]) throws Exception {
Configuration c = new Configuration();
if (args.length != 2) {
System.out.println("provide sufficient arguments");
System.exit(-1);
}
Job job = Job.getInstance(c, "Wordcount");
job.setJarByClass(Wordcount.class);
job.setMapOutputKeyClass(Text.class);
job.setMapOutputValueClass(IntWritable.class);
job.setOutputKeyClass(Text.class);
job.setOutputValueClass(Text.class);
job.setMapperClass(Map.class);
job.setReducerClass(Reduce.class);
job.setInputFormatClass(TextInputFormat.class);
job.setOutputFormatClass(TextOutputFormat.class);
FileInputFormat.addInputPath(job, new Path(args[0]));
FileOutputFormat.setOutputPath(job, new Path(args[1]));
// job.submit();
job.waitForCompletion(true);
}
@Override
public int run(String[] args) throws Exception {
String locatorHost = args[0];
int locatorPort = Integer.parseInt(args[1]);
String hdfsHomeDir = args[2];
System.out.println(
"KnownKeysMRv2 invoked with args (locatorHost = "
+ locatorHost
+ " locatorPort = "
+ locatorPort
+ " hdfsHomeDir = "
+ hdfsHomeDir);
Configuration conf = getConf();
conf.set(GFInputFormat.INPUT_REGION, "partitionedRegion");
conf.set(GFInputFormat.HOME_DIR, hdfsHomeDir);
conf.setBoolean(GFInputFormat.CHECKPOINT, false);
conf.set(GFOutputFormat.REGION, "validationRegion");
conf.set(GFOutputFormat.LOCATOR_HOST, locatorHost);
conf.setInt(GFOutputFormat.LOCATOR_PORT, locatorPort);
Job job = Job.getInstance(conf, "knownKeysMRv2");
job.setInputFormatClass(GFInputFormat.class);
job.setOutputFormatClass(GFOutputFormat.class);
job.setMapperClass(KnownKeysMRv2Mapper.class);
job.setMapOutputKeyClass(GFKey.class);
job.setMapOutputValueClass(PEIWritable.class);
job.setReducerClass(KnownKeysMRv2Reducer.class);
// job.setOutputKeyClass(String.class);
// job.setOutputValueClass(ValueHolder.class);
return job.waitForCompletion(false) ? 0 : 1;
}